OpenClaw MCP Registration Runbook
SkillSkill
Free. The mcpServers block in openclaw.json is silently ignored in v2026.2.26+. This documents the correct path.
About
In OpenClaw v2026.2.26 and later, the mcpServers block in openclaw.json is silently ignored. Your agents do not get the tools and no error is thrown. Anyone trying to register a custom MCP server using the documented method will hit this with no diagnostic signal.
The correct path: servers must be registered in /root/.mcporter/mcporter.json via the mcporter registry layer. This requires a Docker volume mount (/root/.mcporter:/root/.mcporter) in your docker-compose.yml. Without it, the file resets on container restart.
Two additional issues covered: Python MCP servers need an initialize handler or OpenClaw will not detect them at startup. Credentials in server configs should use wrapper scripts reading from /root/secrets/ rather than inline environment variables.
What is included: step-by-step setup instructions, a working mcporter.json example, the Docker volume mount addition for docker-compose.yml, a Python MCP server template with the initialize handler, and verification steps to confirm the server is registered and callable.
Core Capabilities
- Step-by-step mcporter.json setup with working example config
- Docker volume mount pattern to persist MCP registration across restarts
- Python MCP server initialize handler — the missing piece that causes silent failures
- Credential security pattern: wrapper scripts reading from /root/secrets/
- Verification steps: how to confirm your MCP server is registered and callable
- Per-agent tool scoping: restrict specific tools to specific agents
Customer ratings
0 reviews
No ratings yet
- 5 star0
- 4 star0
- 3 star0
- 2 star0
- 1 star0
No reviews yet. Be the first buyer to share feedback.
Version History
This skill is actively maintained.
March 25, 2026
One-time purchase
$0
By continuing, you agree to the Buyer Terms of Service.
Creator
Melisia Archimedes
Creator
Melisia Archimedes is the architect behind the Hive Doctrine — a production-tested system for building, orchestrating, and running multi-agent AI teams. I've spent years in the field, not on the whiteboard. Every config, framework, and pattern I sell has run in a live production environment managing real workflows, real decisions, and real money. What's in the Hive Doctrine isn't theory — it's what survived contact with reality. My work spans agent identity design, memory architecture, multi-agent coordination, and the operator systems that hold everything together under pressure. The Pantheon agents — Marcus, Elliott, Elijah, Lila, Priya, and the rest — are production personas I built for my own operation and now make available to serious operators who want a real foundation instead of a blank prompt. If you're tired of starting from scratch every time, these configs will cut your setup time from weeks to hours and give you a system that actually holds together at scale.
View creator profile →Details
- Type
- Skill
- Category
- Ops
- Price
- $0
- Version
- 1
- License
- One-time purchase
Works great with
Personas that pair well with this skill.
Commercial & Residential Real Estate Intelligence Agent — Elite Claw
Persona
A fully verticalized AI real estate agency in a box. Not a SaaS subscription that treats your business like everyone else. One agent, every department, built 1-of-1 for your operation.
$249
Real Estate Intelligence Agent — Pro Claw
Persona
Not a cheaper version of the software you already hate. A fundamentally different approach to running a real estate business — built from first principles, configured 1-of-1 for your operation.
$129
TokenOps Ninja 🥷
Operations
Agents burn money. Ninja stops them.
$49